本文针对薄膜翼飞行机器人的性能参数优化展开理论分析、结构设计和样机实验。通过设计并制造定量化测试飞控性能的实验台架，研究了协方差矩阵自适应（CMA）进化算法。利用台架实验和算法完善来提高蝙蝠机器人的性能和结构设计。利用进化迭代算法逐步优化蝙蝠机器人的飞行控制参数。这种迭代过程不仅细化了控制参数，而且有助于机器人结构设计的日趋完善。通过结合 CMA 进化策略，实现了蝙蝠机器人的飞行动力学优化，特别是在参数调整方面，到达了机器人与自然蝙蝠在力学性能方面的趋同一致。在此基础上，该研究扩展了进化算法的使用范围，探索并量化了蝙蝠后腿运动对推力和升力产生的影响。值得注意的是，这项研究证实了蝙蝠自然生理的机械效率。蝙蝠机器人的机械参数收敛优化到了自然界中观察值。这些发现不仅验证了进化算法在扑翼机器人设计中的有效性，还增强了我们对飞行哺乳动物生物力学的理解，为未来的仿生机器人系统设计提供了一定的理论、设计和实验基础，也为扑翼机器人的工业应用提供了有启示的借鉴方案。

In this thesis, theoretical analysis, structural design and prototype experiments arecarried out to optimize the performance parameters of the thin-film wing flying robot.The covariance matrix adaptive (CMA) evolutionary algorithm is proposed by designingand manufacturing an experimental bench for quantifying flight control performance. Theperformance and structural design of Batbot are improved by bench experiments and algo-rithm improvement. The flight control parameters of Batbot are optimized by evolutionaryiterative algorithm. This iterative process not only refines the control parameters, but alsocontributes to the improvement of the robot structure design. By combining the CMAevolution strategy, the flight dynamics optimization of the bat robot is realized, especiallyin terms of parameter adjustment. Based on this, the study extends the use of evolution-ary algorithms to explore and quantify the effects of bat hind leg movements on the thrustand lift generated. Notably, the study confirms the mechanical e?iciency of bats’ naturalphysiology. The mechanical parameters of the Batbot are convergent and optimized to theobserved values in nature. These findings not only validate the effectiveness of evolution-ary algorithms in the design of flapping wing robots, but also enhance our understandingof the biomechanics of flying mammals, and provide a certain theoretical, design and ex-perimental basis for the future design of bionic robot systems, but also provide an inspiredreference scheme for the industrial application of flapping wing robots.